32tt project

milagros · Post by **milagros** » 1 Apr 2011, 10:07

Last year I was annoyed that almost noone appreciate my eol effort and send me his wr recs. So I decided to quit eol and use all my hoyling and coding skillz to drive 54 wrs. The project got the name "32tt project" for its target.

Rules were simple :
a) all recs have to be theoretically drivable = they must be result of some time events and key presses on non-xiit elma
b) bounces should have no pop and the strength was limited to "certain" value - some may find some bounces buggy:/
c) minibugs/miniwheelpops like in flattrack wr start allowed
d) bangoing between two edges (like in 31 start) allowed

Not all times were driven with the latest tools yet. In some of them the wr was beaten and they have not been touched again (12,13,14,..). 31tt should be ez, 30tt or even 29tt possible. So here is the list:

Code: Select all

1.  Warm Up              0:13,84
2.  Flat Track           0:14,67
3.  Twin Peaks           0:17,18
4.  Over and Under       0:27,99
5.  Uphill Battle        0:19,77
6.  Long Haul            0:28,59
7.  Hi Flyer             0:23,50
8.  Tag                  0:19,90
9.  Tunnel Terror        0:49,91
10. The Steppes          0:10,69
11. Gravity Ride         0:22,54
12. Islands in the Sky   0:20,63
13. Hill Legend          0:13,85
14. Loop-de-Loop         0:17,55
15. Serpents Tale        0:40,75
16. New Wave             1:07,73
17. Labyrinth            0:51,97
18. Spiral               0:44,16
19. Turnaround           0:34,98
20. Upside Down          0:50,68
21. Hangman              0:14,35
22. Slalom               0:37,96
23. Quick Round          0:33,33
24. Ramp Frenzy          0:35,85
25. Precarious           0:23,88
26. Circuitous           0:19,48
27. Shelf Life           0:32,56
28. Bounce Back          0:33,40
29. Headbanger           0:40,69
30. Pipe                 0:27,63
31. Animal Farm          0:57,20
32. Steep Corner         0:42,30
33. Zig-Zag              0:49,92
34. Bumpy Journey        0:36,19
35. Labyrinth Pro        2:00,02
36. Fruit in the Den     0:56,13
37. Jaws                 0:48,71
38. Curvaceous           0:28,27
39. Haircut              0:37,58
40. Double Trouble       1:17,90
41. Framework            0:36,11
42. Enduro               0:46,25
43. He He                0:52,82
44. Freefall             0:11,35
45. Sink                 0:32,34
46. Bowling              0:46,96
47. Enigma               0:17,04
48. Downhill             0:38,32
49. What the Heck        0:36,99
50. Expert System        0:45,42
51. Tricks Abound        0:33,93
52. Hang Tight           0:23,64
53. Hooked               0:16,88
54. Apple Harvest        1:07,16

Total                   32:29,44

55. More Levels          0:26,79

edit: recs will not be spread ofc

jonsykkel · Post by **jonsykkel** » 1 Apr 2011, 10:23

pawq · Post by **pawq** » 1 Apr 2011, 10:47

naise predikts ;d

Lukazz · Post by **Lukazz** » 1 Apr 2011, 10:50

sick serpents time.

Orcc · Post by **Orcc** » 1 Apr 2011, 11:05

I find the bounce in More Levels semi-legitimate

KD · Post by KD » 1 Apr 2011, 11:18

Grats for the times, they seem pretty good. Hope to see at least 30tt next year same day!

pawq · Post by **pawq** » 1 Apr 2011, 11:24

BRAND NEW
ELASTO MANIA 3D
THE MOST ANTICIPATED PREMIERE OF THE CENTURY
OUT ON 1ST APRIL 2011
EXCLUSIVELY FOR MOPOLAUTA
thx bye

Zweq · Post by **Zweq** » 1 Apr 2011, 11:45

for 31tt you need 1.23.45 in lab pro

Bismuth · Post by **Bismuth** » 1 Apr 2011, 16:12

i return to lauta

milagros · Post by **milagros** » 1 Apr 2011, 18:51

Orcc wrote:I find the bounce in More Levels semi-legitimate

yes I agree, both bounces are way over the top and it would not be accepted elsewhere
http://up.k10x.net/lhrgyldyoleth/55x2679.rec

Lukazz · Post by **Lukazz** » 1 Apr 2011, 19:25

milagros wrote:
Orcc wrote:I find the bounce in More Levels semi-legitimate
yes I agree, both bounces are way over the top and it would not be accepted elsewhere
http://up.k10x.net/lhrgyldyoleth/55x2679.rec

... the fuck!

jonsykkel · Post by **jonsykkel** » 1 Apr 2011, 19:29

heh nice one, ez better with my style tho
http://up.k10x.net/iqkokbszjndpq/55jon1557.rec

milagros · Post by **milagros** » 1 Apr 2011, 19:38

jonharkulsykkel wrote:heh nice one, ez better with my style tho
http://up.k10x.net/iqkokbszjndpq/55jon1557.rec

wonder if it's driven/drivable or some xiiit
edit: watched it again, i bet its not driven, it does not bounce when wheel pops that direction

NoZKeY · Post by **NoZKeY** » 1 Apr 2011, 23:41

jonharkulsykkel wrote:heh nice one, ez better with my style tho
http://up.k10x.net/iqkokbszjndpq/55jon1557.rec

propro, but ez better

Madness · Post by **Madness** » 2 Apr 2011, 19:27

ez sec faster in steep corner

Lousku · Post by **Lousku** » 14 Apr 2011, 11:54

I like your nick, Gomsturlsteeft. One day I will be a spambot too.

Memphis · Post by **Memphis** » 18 Apr 2011, 19:52

U got recs for all of them mila?

milagros · Post by **milagros** » 18 Apr 2011, 21:14

milagros · Post by **milagros** » 28 Apr 2011, 23:43

[6:19pm] <Madness__> mila when update 32 tt project ?
[6:28pm] <ribot> well who is to believe those recs are real
[6:30pm] <Madness__> it is real, and everybody who didnt believe was fooled ;o

clever guy

Ecchi · Post by **Ecchi** » 2 May 2011, 07:57

I just have to ask, did you use any tools to find the styles and shit or is that just pure brainwork?

trew · Post by **trew** » 2 May 2011, 09:37

i find it hard to believe that mila made a tool that invent styles.

pawq · Post by **pawq** » 2 May 2011, 10:08

Either sick brains or sick coding skills :)
Why do you say that there are plenty of "possible" styles that will never be done mila? Isn't it just about reaching the sufficient perfectness? And isn't that possible too?

Kopaka · Post by **Kopaka** » 2 May 2011, 12:39

Pawq wrote:Either sick brains or sick coding skills
Why do you say that there are plenty of "possible" styles that will never be done mila? Isn't it just about reaching the sufficient perfectness? And isn't that possible too?

possible and will be done is not the same

with current amount of wrs it doesn't exactly seem like a lot of new stuff will be done

milagros · Post by **milagros** » 2 May 2011, 12:44

pipe brutal style is right at the beginning, known for 10 years, tried a lot and even though it is perfectly possible, noone managed to do it (even with normal s/l)
such styles are often needed in the middle of the lev, or possible 2 or 3 of such tricks are required to finish the run
they will all not be ever done and won't even be properly tried
edit: updated times now that i'm here:)

Bjenn · Post by **Bjenn** » 2 May 2011, 12:57

You gotta hev sick brains to have sick coding skills, anyway it's mega nice project i'm very impressed that this is possible.

nick-o-matic · Post by **nick-o-matic** » 2 May 2011, 15:46

Björn wrote:You gotta hev sick brains to have sick coding skills, anyway it's mega nice project i'm very impressed that this is possible.

I don't know how it's done but I bet that major part is done with pure driving and stylefinding skills because I think that a program that could really find radically new styles would be very heavy. I believe it's saveload + some program that optimizes the moves with try and error and gives maximum strength to bounces. So I'd salute the hoyling part too, not only the programming part. :)

milagros wrote:edit: recs will not be spread ofc

I guess that 48 and 55 were just for a proof that you really have done this right? I hope so because I think that spreading (more) recs would indeed not be good. :)

trew · Post by **trew** » 2 May 2011, 15:55

nick-o-matic wrote:
Björn wrote:You gotta hev sick brains to have sick coding skills, anyway it's mega nice project i'm very impressed that this is possible.
I don't know how it's done but I bet that major part is done with pure driving and stylefinding skills because I think that a program that could really find radically new styles would be very heavy. I believe it's saveload + some program that optimizes the moves with try and error and gives maximum strength to bounces. So I'd salute the hoyling part too, not only the programming part.
milagros wrote:edit: recs will not be spread ofc
I guess that 48 and 55 were just for a proof that you really have done this right? I hope so because I think that spreading (more) recs would indeed not be good.

and we didn't really get to see the 48 one... it was more of a teaser.

Memphis · Post by **Memphis** » 2 May 2011, 16:04

I don't see an issue with spreading video of the recs. It's sounds like too cool of a program to keep viewing only to yourself.

J-sim · Post by **J-sim** » 2 May 2011, 18:09

lol so the joke is actually on us who were sure this was an april fools' joke?!

well would've been too much of a copy of abula's one if it had been.

Lukazz · Post by **Lukazz** » 2 May 2011, 20:57

Memphis wrote:I don't see an issue with spreading video of the recs. It's sounds like too cool of a program to keep viewing only to yourself.

i agree very much. recs shouldn't be shared of course, but someone could make "elma xiited quick" or something, that would be awesome.

nick-o-matic · Post by **nick-o-matic** » 2 May 2011, 20:59

I think that would ruin the joy of discovering new styles and making surprising WRs for forever.

Lukazz · Post by **Lukazz** » 2 May 2011, 21:57

nick-o-matic wrote:I think that would ruin the joy of discovering new styles and making surprising WRs for forever.

i don't think so. internal scene is almost dead anyway, and i think a few new styles could give it a last little push, to pull 34 wr TT of and then we can forget about the whole internal thing forever.

Lousku · Post by **Lousku** » 3 May 2011, 10:47

What's bad about sharing the recs?

I'm asking this as a nab who is totally tired of the internal scene because of jealousy.

Bjenn · Post by **Bjenn** » 3 May 2011, 10:50

Share all please or better, make a vid, nothing bad will happen, only good. And my motivation will raise aswell for internals, after watching the 48 vid i played int 37 more, that's a proof.

trew · Post by **trew** » 3 May 2011, 14:25

Lousku wrote:What's bad about sharing the recs?

I'm also a little interested why it is bad to share the recs? If it's just that mila doesn't want to, I'm fine with that. but what else?

Igge · Post by **Igge** » 3 May 2011, 14:29

What if Zweq's hi flyer and ramp frenzy had been shared as saveload recs before table 265?

Na. I want styles to be found through actual WRs, not through cheats.

trew · Post by **trew** » 3 May 2011, 14:46

Igge wrote:What if Zweq's hi flyer and ramp frenzy had been shared as saveload recs before table 265?

Na. I want styles to be found through actual WRs, not through cheats.

Maybe so, but there are obviously styles that will never be WRs, and therefore cannot be shared that way.

Grace · Post by **Grace** » 3 May 2011, 15:26

I'm interested to see some of the recs (warm up, Anal farm, Upside down etc) Sick project

Memphis · Post by **Memphis** » 3 May 2011, 15:48

but that's the thing. If shared the recs it would turn into a man versus machine competition and if man beat any of Mila's recs, technically there is a flaw in the program.

I think sharing the recs would be good for the scene or make a vid of them. That's my 2 cents. Internal scene is struggling as is like bjorn said. Not many wrs so I think doing this would speed up progress and who knows, maybe still a style or two to be found even after recs are released.

Kiiwi · Post by **Kiiwi** » 3 May 2011, 16:08

Whole project is just shit...sharing the recs even more so....just my opinion

And before anyone starts barking at me i'll tell why i think so: Based on the released vids the styles and tricks are completely out of reach in one run, if everyone could use saveload and whatnot tools, then fine, everyone could compete who gets 32tt or whatever first...now it's just showing off and telling us how we suck at stylefinding and how everyone of these tricks are doable....well i'd like to see someone make, let's say, that hiflyer time or any of those released vids......yeah right, maybe zweq makes some 40 in downhill but that's it.

milagros · Post by **milagros** » 3 May 2011, 16:10

Memphis wrote:if man beat any of Mila's recs, technically there is a flaw in the program

wrong, it just means I wasn't able to find the same style (no idea what's Kazan's trix, if he wants to send me) or didn't manage to reproduce with my tools
there are many sucky times atm which can be beaten "easily", especially those quick 0.02 improvements
new styles can be still found, for example romy4's sl rec (or who made it) in haircut used new style and is only some 0.2s behind my rec after start, so it may possibly be faster, haven't tried yet
as i mentioned, ez 31.xx if i keep playing, or ez 29.xx if teh tools get spread

milagros · Post by **milagros** » 3 May 2011, 16:14

Kiiwi wrote:now it's just showing off

yeah
and about those styles.. I personally would not be able to beat any WR in any lev even by knowing styles possible (maybe in 02 or 44, or maybe not)
there are only some 4-5 styles which would possibly lead to new wrs
most of them are sick fucks like those 2 spread, or just hoyled

Memphis · Post by **Memphis** » 3 May 2011, 16:40

Ya it's not like this program finds every new style out there. It sounds like mila needs to know the basis of the style and then enter it in the program. Recs shared is a good idea. Especially Warm up!

culinko · Post by **culinko** » 3 May 2011, 18:42

milagros wrote:there are only some 4-5 styles which would possibly lead to new wrs

you invented them or the program itself?

Lousku · Post by **Lousku** » 3 May 2011, 18:56

CuLiNo wrote:you invented them or the program itself?

Yeah, mila built a machine that churns out WR styles and eventually recs when you feed it potatoes and push a button.

trew · Post by **trew** » 3 May 2011, 19:29

Lousku wrote:
CuLiNo wrote:you invented them or the program itself?
Yeah, mila built a machine that churns out WR styles and eventually recs when you feed it potatoes and push a button.

isn't he awesome?

culinko · Post by **culinko** » 3 May 2011, 20:37

Lousku wrote:
CuLiNo wrote:you invented them or the program itself?
Yeah, mila built a machine that churns out WR styles and eventually recs when you feed it potatoes and push a button.

that's not what i've asked.

Lousku · Post by **Lousku** » 3 May 2011, 20:44

CuLiNo wrote:
Lousku wrote:
CuLiNo wrote:you invented them or the program itself?
Yeah, mila built a machine that churns out WR styles and eventually recs when you feed it potatoes and push a button.
that's not what i've asked.

Ah, sorry, misunderstood you. But he did that anyway. Yes, he is awesome.

anpdad · Post by **anpdad** » 4 May 2011, 06:20

program that invents styles? well, control theory could provide neat results in a simplified modelled engine with fixed low fps, in a reasonable time. With recs. Even more simplified version without any physics of the bike could be applicable too, to simply outline the possible routes. This would be cool to see, i wonder if he used that, or something more primitive instead. I'm still convinced none of that was in use though. Brain styles, saveload driving, optimize bounces with proggy, ...., profit!

Lousku · Post by **Lousku** » 4 May 2011, 08:57

Does someone seriously think milagros made a style-inventing höyläbot?

anpdad wrote:Brain styles, saveload driving, optimize bounces with proggy, ...., profit!

This is obvious.

But the program does run on potatoes.